Transmission Design And Optimization In Cooperative Multi-user Networks

Cooperation technique in wireless communication makes it possible for single antenna nodes to share each other's resource. This will form a virtual antenna array and can effectively reduce multipath fading. In multi-user cooperation, not only the primary user communicates with the base station, but there are also vacant users assisting the communication as cooperative nodes. This paper studies non-cognitive as well as cognitive cooperative multi-user networks, and investigates on transmission design and optimization. The main contributions are as follows:(1) For non-cognitive cooperative multi-user networks, we propose a novel Compressed Sensing (CS) based amplify-and-forward coding scheme, which utilizes CS as a joint source-channel code. It is the first time when CS is used in amplify-and-forward channels as for coding. Our coding scheme outperforms the existing CS based decode-and-forward coding scheme.(2) For cognitive cooperative multi-user networks, we propose four different transmission schemes and formulates the system optimization problems. Specifically, we propose a compressive joint coding scheme when secondary users (SUs) have perfect channel state information at the transmitters (CSIT). This scheme is very suitable for cognitive networks by using a rateless channel code, and improves robustness in a suddenly deteriorated channel due to the protection ability of CS as a channel code.When the primary and secondary sources are correlated and SUs have perfect CSIT, we propose a hybrid digital-analog coding scheme, which first applies hybrid digital-analog coding in an overlay cognitive radio network. Our scheme is better than existing schemes for canceling the interference from the SU to the primary user (PU). Moreover, we propose a transmission scheme based on linear assignment Gel'fand-Pinsker coding (LA-GPC) when SUs only have partial CSIT, solving the problem that dirty paper coding (DPC) will introduce a significant performance loss in this case. Our scheme eliminates the SU's interference to the PU.Considering a cognitive radio network with multiple relays, we propose a relaying scheme which satisfies the constraint on SUs in underlay mode,and can allocate maximum power to the SU while ensuring the PU's priority.All transmission schemes mentioned above are provided with simulation results. The transmission designs and theoretical derivations are proved finally.